Many vehicles make use of anti-collision systems to help reduce accidents between vehicles and hazards such as pedestrians, road obstructions, or other vehicles. These systems use interconnected sensor networks such as radar, optical sensors, infrared, or laser sensors to detect and track objects in the environment surrounding the vehicle.
While current anti-collision systems have become proficient and reliable in identifying objects like other vehicles and road obstructions, tracking pedestrians has proven more difficult. As collisions between vehicles and pedestrians are a very serious problem, the need exists for anti-collision systems that are capable of reliably identifying and tracking pedestrians.
Optical based sensors such as cameras usually can provide high-resolution images and wide observation field. However, these systems often require substantial computing and data processing to extracting target information. In addition, optical based systems are often limited by weather and lighting conditions.
While infrared technology can overcome some of the deficiencies of optical based sensors, these systems are still limited with respect to short range detection. Laser based sensor networks can provide accurate information regarding a target's position and distance. However, laser networks continue to have safety issues due to the lack of textured information about the targets.
Radar systems are well known in the art for measuring both distance and relative speed of objects surrounding vehicles. These systems can be used to improve the driver's ability to perceive objects when visibility is poor or in hard to see areas such as blind spots. Recently, the 76-78 GHz radar frequency band was allocated for automobile collision avoidance radar systems. Due to its longer detection range, better position resolution, and reduced sensitivity to various road and weather conditions, there is significant research and development on radar systems in this frequency range.
Compared to optical sensor systems, 76-78 GHz radar has the advantages of longer detection range, higher target position resolution, and better tolerance for poor weather and lighting conditions. In addition, radar data contains Doppler and micro Doppler signatures associated with motions of different moving parts of a target. This makes it ideal for pedestrian collision warning/avoidance systems on vehicles. Testing the effectiveness of such systems requires the use of special pedestrian mannequins that produce a radar response similar to a real human from different observation angles and in different postures.
The size and shape of an object are two of the factors that influence the response that the object will generate when exposed to a radar signal. Additionally, different surface materials produce different responses to 76-78 GHz radar. Therefore, to accurately test pedestrian collision systems, radar mannequins not only need to have body shapes that resemble real pedestrians, but also artificial skins that replicate the radar reflectivity of real human skin. In addition, the radar response generated by a real person is not consistent throughout their entire body. The response generated by a torso is different than the response generated by a lower leg. As such, the radar mannequin should be configurable to produce a non-uniform radar response.
It would therefore be beneficial to produce an artificial skin specifically designed for radar mannequins used in 76-78 GHz vehicular collision avoidance radar evaluations and testing. The artificial skin should enable a radar mannequin to produce radar cross section (RCS) pattern data that closely resembles that of a real person. A radar cross section is a measure of how detectable an object is with radar. The artificial skin is configured so that the radar mannequin, from the perspective of a radar system, appears like a human. In addition, as the radar response of a human can vary from one body part to another, the artificial skin should be easily configurable so that the response of specific body parts of the radar mannequin can match their real human counterparts.